KR20110126075A - Method and apparatus for video encoding and decoding using extended block filtering - Google Patents

Abstract

PURPOSE: A method and apparatus for performing video encoding and decoding using an extended block filtering are provided to create the corresponding block of the reference picture into an interpolated reference picture. CONSTITUTION: A block extension unit increase the size of the reference picture to a predetermined size(710). A first filtering part applies the first filter to the extended corresponding block(720). A second filtering part applies a predetermined second filter to the filtered expansion corresponding block(730). A motion prediction unit and a motion correction unit perform a motion estimation and compensation in a sub pixel unit based on the interpolated reference picture.

Description

Method and apparatus for video encoding and decoding using extended block filtering

The present invention relates to an interpolation filter for interpolating a reference image for motion prediction and compensation. More particularly, the present invention relates to an interpolation filter that extends a corresponding block of a reference picture to perform filtering by applying an infinite impulse response filter (IIR). Next, the present invention relates to a video encoding method and apparatus for performing interpolation on a subpixel basis using an IIR filtered extended block, and a decoding method and apparatus thereof.

In video compression schemes such as MPEG-1, MPEG-2, MPEG-4, and H.264 / MPEG-4 Advanced Video Coding (AVC), a picture is divided into macro blocks to encode an image. Each macroblock is encoded in all encoding modes available for inter prediction and intra prediction, and then one encoding mode is selected according to the bit rate required for encoding the macro block and the degree of distortion of the original macro block and the decoded macro block. Select to encode the macro block.

Inter prediction generates a motion vector by searching an area of a reference picture similar to the current block encoded using at least one reference picture located in front of or behind the currently encoded picture, and performs motion compensation using the generated motion vector. To generate a prediction block of the current block.

Motion compensation is performed in sub-pixel resolution units more precise than integer pixel resolution to improve prediction accuracy. For example, a subpixel such as half pel, quarter pel, and one eighth pel between integer pixels of a reference picture is generated to interpolate the reference picture. Then, motion compensation is performed using the interpolated reference picture.

The technical problem to be solved by the present invention is to extend the corresponding block of the reference picture used for motion compensation and to generate interpolated reference pictures by applying block-based infinite impulse response filtering and finite impulse response filtering to predict the efficiency of motion compensation of video. To provide a video encoding method and apparatus, and a decoding method and apparatus.

In addition, a technical problem of the present invention is to apply a filter having an infinite impulse response only to a predetermined region of a reference picture, a video encoding method and apparatus for improving prediction efficiency without greatly increasing the complexity of the operation, and a decoding method thereof To provide a device.

According to an aspect of the present invention, there is provided a video encoding method comprising: generating an extended corresponding block by extending a corresponding block of a reference picture used for motion compensation of a current block to a predetermined size; Generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block; Performing interpolation on a sub-pixel basis by applying a second filter to the filtered extended corresponding block; And performing motion prediction and compensation using the reference picture interpolated in the sub-pixel units.

A video decoding method according to an embodiment of the present invention comprises the steps of extracting motion vector information of a current block to be decoded from the received bitstream; Generating an extended corresponding block by extending a corresponding block of a reference picture indicated by the motion vector to a predetermined size; Generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block; Performing interpolation on a sub-pixel basis by applying a second filter to the filtered extended corresponding block; And performing motion compensation using the reference picture interpolated in the sub-pixel units.

A video encoding apparatus according to an embodiment of the present invention includes a block extension unit for generating an extended corresponding block by extending a corresponding block of a reference picture used for motion compensation of a current block to a predetermined size; A first filtering unit generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block; A second filtering unit configured to perform interpolation on a subpixel basis by applying a predetermined second filter to the filtered extended corresponding block; And a motion prediction and compensation unit configured to perform motion prediction and compensation using the reference picture interpolated in the sub-pixel units.

An apparatus for decoding video according to an embodiment of the present invention includes an entropy decoding unit for extracting motion vector information of a current block to be decoded from a received bitstream; A block extender configured to expand the corresponding block of the reference picture indicated by the motion vector to a predetermined size to generate an extended corresponding block; A first filtering unit generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block; A second filtering unit configured to perform interpolation on a subpixel basis by applying a predetermined second filter to the filtered extended corresponding block; And a motion compensator for performing motion compensation using the reference picture interpolated in the sub-pixel units.

According to the present invention, it is possible to improve compression efficiency in motion prediction and compensation of video.

1 is a block diagram illustrating a configuration of a video encoding apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a detailed configuration of the reference picture interpolation unit 170 of FIG. 1.
3 is a reference diagram for explaining a process of generating an extension corresponding block according to an embodiment of the present invention.
4 is a reference diagram for describing a filtering operation performed by the first filtering unit 220 in a row direction according to an embodiment of the present invention.
5 is a reference diagram for describing a filtering operation performed by the first filtering unit 220 in a column direction according to an embodiment of the present invention.
6 is a reference diagram for explaining an interpolation process performed by the second filtering unit 230 of FIG. 2.
7 is a flowchart illustrating a video encoding method according to an embodiment of the present invention.
8 is a flowchart illustrating a video decoding apparatus according to an embodiment of the present invention.
9 is a flowchart illustrating a video decoding method according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a configuration of a video encoding apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the video encoding apparatus 100 according to an embodiment of the present invention may include a motion predictor 110, a motion compensator 120, a subtractor 130, an encoder 140, and a reconstructor ( 150, a storage unit 160, and a reference picture interpolator 170.

The motion predictor 110 generates a motion vector of the current block by performing motion prediction using data of the reference picture. In detail, the motion predictor 110 searches for a corresponding block of a reference picture most similar to the current block to be encoded of the current picture, and generates a motion vector based on a position difference between the current block and the corresponding block. In this case, the motion predictor 110 generates a motion vector in units of integer pixels by using a reference picture in units of integer pixels stored in the storage unit 160 when first predicting motion. In addition, the motion predictor 110 may use not only a reference picture in an integer pixel unit as a reference picture but also a reference picture interpolated at a subpixel resolution such as 1/2 pixel, 1/4 pixel, etc. by the reference picture interpolator 170. Motion prediction can be performed.

The reference picture interpolator 170 expands the corresponding block of the reference picture indicated by the integer pixel motion vector based on the integer pixel motion vector of the current block to generate an extended corresponding block, and generates an infinite impulse response to the extended corresponding block. A filtered extended response block is generated by applying a first filter having an Infinite Impulse Response (IIR) characteristic. In addition, the reference picture interpolator 170 generates a reference picture interpolated at sub-pixel resolution by applying a second filter having a finite impulse response (FIR) characteristic to the filtered extended corresponding block. The detailed operation of the reference picture interpolator 170 will be described later.

The motion compensator 120 generates a prediction block of the current block by obtaining a motion compensation value of the current block indicated by the motion vector of the current block from the interpolated reference picture. If the determined optimal motion vector is an integer pixel unit, the motion compensator 120 may perform motion compensation using the non-interpolated reference picture without using the interpolated reference picture.

The subtractor 130 calculates a residual block that is a difference value between the prediction block and the original input block. The encoder 140 generates a bitstream by transforming, quantizing, and entropy encoding the residual block. In addition, the encoder 140 compensates for the motion by using the interpolated reference picture by filtering the corresponding block extended by the reference picture interpolator 170 together with the motion vector information of the current block in a predetermined region of the generated bitstream. Predetermined index information indicating whether this has been performed may be inserted. That is, according to the prior art, when the reference picture is interpolated, '0' is used, and when the reference picture is interpolated using the filtering of the extended corresponding block according to an embodiment of the present invention, one bit having a value of '1' is used. By inserting a flag into the bitstream, it is possible to determine whether to generate an interpolated reference picture using filtering of the extended corresponding block at the decoding end. Such index information may be set in a slice unit and a sequence unit and included in a slice header or a sequence header.

FIG. 2 is a block diagram illustrating a detailed configuration of the reference picture interpolation unit 170 of FIG. 1.

Referring to FIG. 2, the reference picture interpolator 170 includes a block extension 210, a first filtering unit 220, and a second filtering unit 230.

The block extension unit 210 generates an extended corresponding block by extending the corresponding block of the reference picture indicated by the motion vector of the current block determined by the motion predictor 110 to a predetermined size.

The first filtering unit 220 generates a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block. The second filtering unit 230 performs interpolation in units of subpixels by applying a predetermined second filter to the filtered extended corresponding block.

3 is a reference diagram for explaining a process of generating an extension corresponding block according to an embodiment of the present invention.

2 and 3, the block extension unit 210 corresponds to the corresponding block 321 of the reference picture 320 by using the integer pixel unit motion vector MV_INT of the current block 311 of the current picture 310. Can be determined. That is, when the motion prediction unit 110 performs the motion prediction in units of integer pixels and determines the motion vector MV_INT in units of integer pixels, the block extension 210 indicates the motion vector MV_INT in units of the determined integer pixels. The boundary of the corresponding block 321 of the reference picture 320 is extended by a predetermined pixel size to generate an extended corresponding block 322. Specifically, when the size of the current block 311 is XxY (where X and Y are integers), the block expansion unit 210 crosses the boundary of the corresponding block 321 of the reference picture 320 in a horizontal direction, Expanding by b pixels in the longitudinal direction, produces an extended correspondence block 322 having a size of (X + 2a) x (Y + 2b). Here, the block extension unit 210 may determine how much to extend the corresponding block 321 according to the size of the current block 311 to be encoded. For example, the values of a and b may have values of 1, 3, and 6 in proportion to the size of the current block 311. That is, assuming that the current block 311 may have sizes of 16x16, 8x8, and 4x4, when the current block 311 is 16x16, the block extension 210 may set the upper, lower, left, and right boundaries of the corresponding block 321 to 6, respectively. If the current block 311 is 8x8, the block extension unit 210 expands the upper, lower, left, and right boundaries of the corresponding block 321 by 3 pixels. If the current block 311 is 4x4, the block extension unit 210 expands the upper, lower, left, and right boundaries of the corresponding block 321 by 1 pixel unit, and expands it by 6x6. An expansion corresponding block 322 of size may be generated.

The first filtering unit 210 generates the filtered extended corresponding block by applying the first filter having the IIR characteristic to the extended corresponding block 322. The IIR filter is a filter in which a value of an input signal and a value of an output signal are applied recursively, and refers to a filter having an infinite length of an impulse response as a characteristic function. Since it is impossible to actually perform an infinite calculation to calculate the output value of the IIR filter, an IIR filter is generally applied on a frame-by-frame basis in image processing. However, when the IIR filter is applied frame by frame, the compression efficiency of the image can be improved, but the complexity of the operation increases because the IIR filter must be applied to the entire frame. Accordingly, one embodiment of the present invention applies the IIR filter based on the extended corresponding block to reduce the complexity of the operation while maintaining the efficiency of the IIR filter.

In an embodiment, the first filtering unit 210 may be implemented through a filter having a transfer function characteristic as shown in Equation 1 in the z-domain.

In Equation 1, c1 and z1 are predetermined constant values. Equation 1 is expressed using the recursive algorithm as follows.

In Equation 2, s (k) represents the k-th sample value in the row or column direction to be filtered. Referring to Equation 2, c + (k) is a causal filter using the (k-1) th result before the k th sample value currently filtered, and c- (k) is the k th sample value currently filtered. This is a non-causal filter that uses the (k + 1) th result. The first filtering unit 220 generates a filtered extended corresponding block by performing one-dimensional filtering on the extended corresponding block 322 in a row direction and a column direction by using a recursive equation such as Equation 2. In the above-described embodiment, the first filtering unit 220 uses an IIR filter having the transfer function characteristics as shown in Equations 1 and 2, but the transfer function characteristics of the IIR filter performed in the first filtering unit 220 are different recursion. It will be understood by those of ordinary skill in the art that the present invention can be implemented through an adaptive algorithm.

4 is a reference diagram for describing a filtering operation performed by the first filtering unit 220 in a row direction according to an embodiment of the present invention, and FIG. 5 is a first diagram in a column direction according to an embodiment of the present invention. A reference diagram for describing a filtering operation performed by the filtering unit 220 is described.

Referring to FIG. 4, the size of the extended correspondence block 400 is NxM (N, M is an integer), z1 is a predetermined constant, and the k th (k is 0 to N−) located in a predetermined row 410 of the extended corresponding block. Integer up to 1) When a pixel value of an integer pixel is s1 (k), the first filtering unit 220 includes the following equation based on Equation 2 described above; The value of c2 (k) calculated using c1 (k) = s1 (k) + z1 * c1 (k-1) and c2 (k) = z1 * (c2 (k + 1) -c1 (k)) By using s1 (k) by replacing the value to perform the one-dimensional filtering of the row direction for the extended corresponding block. The first filtering unit 220 performs the one-dimensional filtering operation in the row direction on all the rows of the extended correspondence block 400.

Referring to FIG. 5, the pixel value of the k th integer (k is an integer from 0 to M-1) located in a predetermined column 510 of the extended correspondence block 500 which is filtered in the row direction is s2 (k). In this case, the first filtering unit 220 includes the following equation; The value of c4 (k) calculated using c3 (k) = s2 (k) + z1 * c3 (k-1) and c4 (k) = z1 * (c4 (k + 1) -c3 (k)) 1D filtering in the column direction for the extended correspondence block 500 is performed by substituting the value of s2 (k) using. As described above, the first filtering unit 220 has an integer pixel that is finally filtered by applying an IIR filter in a row direction and filtering the result value again in the column direction by applying an IIR filter having the same filter characteristic. Outputs an extended correspondence block.

Referring back to FIG. 2, when the filtering for each integer pixel included in the extended corresponding block is completed by applying the IIR filter to the corresponding block extended by the first filtering unit 220, the second filtering unit 230 is performed. Performs interpolation on a sub-pixel basis using an FIR filter and an average value filter of a predetermined tap coefficient using IIR filtered integer pixel values.

6 is a reference diagram for explaining an interpolation process performed by the second filtering unit 230 of FIG. 2. 6 illustrates a positional relationship between the interpolated signal in sub-pixel units and the integer pixels filtered by the first filtering unit 220.

Referring to FIG. 6, uppercase letters A through U represent integer pixels filtered by the first filtering unit 220. Lowercase letters a through s shown in the middle represent interpolated pixels with 1/2 pixel precision and 1/4 pixel precision.

The second filtering unit 230 applies the FIR filter having a predetermined tap coefficient to the integer pixels filtered by the first filtering unit 220 to generate a prediction signal having 1/2 pixel precision. For example, the second filtering unit 230 transmits a 6-tap FIR filter having a tap coefficient of {(1, -5, 20, 20, -5, 1) / 32} to the first filtering unit 220. The interpolation is performed on the integer pixels filtered by the interpolation pixels, and the pixels having the 1/4 pixel precision are interpolated by using a 2-tap average filter. As an example, the second filtering unit 230 may apply horizontal 6-tap filtering to the integer pixels E, F, G, H, I, and J filtered by the first filtering unit 220, thereby providing 1/2 pixel precision. A value of 1/2 pixel b with is generated using the following equation; b = (E-5F + 20G + 20H-5I + J) / 32. In addition, the second filtering unit 230 performs 6-tap filtering in the vertical direction to generate 1/2 pixel positioned in the vertical direction of two integer pixels. For example, the second filtering unit 230 converts 1/2 pixel h into the following equation; generated using h = (A-5C + 20G + 20M-5R + T) / 32. The second filtering unit 230 generates 1/2 pixels positioned between four integer pixels by applying 6-tap filtering in both horizontal and vertical directions. For example, in the case of 1/2 pixel j, the second filtering unit 230 first generates 1/2 pixel signals aa, bb, b, j, s, gg and hh by horizontal 6-tap filtering. , Again by applying vertical six-tap filtering, i.e. Generate 1/2 pixel j via j = (aa-5bb + 20b + 20s-5gg + hh) / 32. Alternatively, the second filtering unit 230 first generates 1/2 pixels cc, dd, h, m, ee, and ff by filtering in the vertical direction, and then the following equation; We can generate a white 1/2 pixel j for horizontal filtering, such as j = (cc-5dd + 20h + 20m-5ee + ff) / 32.

The second filtering unit 230 may generate subpixels having a 1/2 pixel precision and then generate 1/4 pixels using an average filter. 1/4 pixels a, c, i, k are generated using a horizontal mean filter of integer pixels or 1/2 pixel adjacent to the periphery. For example, the second filtering unit 230 may be represented by the following equation; Generate a quarter pixel a using an average filter such as a = (G + b). Similarly, 1/4 pixels d, f, n, q are generated using a vertical mean filter of integer pixels or 1/2 pixel adjacent to the periphery. For example, the second filtering unit 230 generates 1/4 pixel f by using an average filter such as f = (b + j) / 2. A diagonal average filter may be used for 1/4 pixels e, g, p, and r. For example, the second filtering unit 230 may be represented by the following equation; Generate a 1/4 pixel r using an average filter such as r = (m + s) / 2.

As described above, when the reference picture interpolated in units of subpixels is generated by the second filtering unit 230, the motion predictor 110 searches for a reference picture area most similar to the current block by using the interpolated reference picture. A motion vector with sub pixel precision can be determined. When the final motion vector is determined by comparing rate-distortion (RD) costs according to each motion vector, motion compensation is performed using the interpolated reference picture by using information about the value of the motion vector and filtering of the extended corresponding block. Predetermined index information indicating whether or not the information has been performed may be inserted into the encoded bitstream as encoding information.

7 is a flowchart illustrating a video encoding method according to an embodiment of the present invention.

In operation 710, the block extension unit 210 expands the corresponding block of the reference picture used for motion compensation of the current block to a predetermined size to generate an extended corresponding block. As described above, the block extension unit 210 determines the corresponding region of the reference picture based on the motion vector of the integer pixel unit determined by the motion predictor 110, and then corresponds to the corresponding block of the current block in proportion to the size of the current block. Can be extended.

In operation 720, the first filtering unit 220 generates a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block. As described above, the first filtering unit 220 outputs an extended corresponding block having a changed pixel value by performing filtering by applying an IIR filter based on each integer pixel included in the extended corresponding block.

In operation 730, the second filtering unit 230 performs interpolation in units of subpixels by applying a predetermined second filter to the filtered extended corresponding block. As described above, the second filtering unit 230 applies a FIR filter and an average value filter having a predetermined tap coefficient based on integer pixels of the extended correspondence block filtered by the first filtering unit 220 to be 1/2. An interpolated reference picture is generated by generating subpixel values with pixel or quarter pixel precision.

In operation 740, the motion predictor 110 and the motion compensator 120 perform motion prediction and compensation on a sub-pixel basis based on the interpolated reference pictures.

8 is a flowchart illustrating a video decoding apparatus according to an embodiment of the present invention.

Referring to FIG. 8, the video decoding apparatus 800 according to an embodiment of the present invention includes an entropy decoder 810, a reordering unit 815, an inverse quantization unit 820, an inverse transform unit 825, and an adder 830. ), A reference picture interpolator 835, a motion compensator 840, and an intra predictor 845.

The entropy decoder 810 receives the compressed bitstream to perform entropy decoding to generate quantized coefficients, while using motion vector information of the current block to be decoded and filtering of the extended corresponding block according to an embodiment of the present invention. By using the interpolated reference picture, predetermined index information indicating whether motion compensation is performed or the like is extracted. The reordering unit 815 reorders the quantized coefficients, and the inverse quantization unit 820 and the inverse transform unit 825 restore the residual by performing inverse quantization and inverse transformation on the quantized coefficients.

The reference picture interpolator 835 interpolates the reference picture for motion compensation of the current block, which is decoded similarly to the reference picture interpolators 170 and 200 described above with reference to FIGS. 1 and 2. In detail, when the motion vector of the current block is configured by an integer pixel unit, the reference picture interpolation unit 170 does not perform interpolation because the previously reconstructed reference picture may be used without any extra interpolation process. However, while the motion vector of the current block has motion in sub-pixel units such as 1/2 pixel or 1/4 pixel, motion compensation is performed by using an interpolated reference picture using filtering of a corresponding block in which a predetermined index is extended. In this case, the reference picture interpolator 835 first determines the corresponding region of the reference picture by using the integer pixel unit component of the motion vector of the current block. For example, when the motion vector of the current block is (4.25, 0.75), the reference picture interpolator 835 determines a corresponding region of the reference picture by using an integer pixel unit motion vector of (4,0). The reference picture interpolator 835 expands the corresponding block of the reference picture used for motion compensation of the current block to a predetermined size to generate an extended corresponding block. In addition, similar to the reference picture interpolator 200 of FIG. 2, the reference picture interpolator 835 of the decoding apparatus 800 generates an filtered extended corresponding block by applying an IIR filter to the extended corresponding block. Next, interpolation is performed in units of subpixels by applying an FIR filter and an average value filter having a predetermined tap coefficient to the IIR filtered extended corresponding block according to the subpixel resolution of the current block.

The motion compensator 840 performs motion compensation by using the interpolated reference picture in subpixel units. The adder 830 restores the current block by adding the motion compensation value of the current block and the restored residual. If the current block is intra predicted, the predicted block predicted by the intra predictor 845 and the reconstructed residual are added to reconstruct the current block.

9 is a flowchart illustrating a video decoding method according to an embodiment of the present invention.

In step 910, the entropy decoder 810 extracts motion vector information of the current block to be decoded from the received bitstream. The motion vector information may include predetermined index information indicating whether to perform motion compensation by using the motion vector value of the current block to be decoded and the reference picture interpolated by filtering the extended corresponding block.

In operation 920, the reference picture interpolator 835 expands the corresponding block of the reference picture indicated by the motion vector to a predetermined size to generate an extended corresponding block. As described above, the reference picture interpolator 835 may determine a corresponding region of the reference picture by using a motion vector component of an integer pixel unit.

In operation 930, the reference picture interpolator 835 generates a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block. The first filter is an IIR filter and is not applied on a frame basis but is applied to an extended corresponding block as described above.

In operation 940, the reference picture interpolator 835 performs interpolation on a subpixel basis by applying a predetermined second filter to the filtered extended corresponding block. Here, the second filter may be a 6 tap filter having a predetermined tap coefficient, a 4 tap filter, a 2 tap filter which is an average value filter, and the like as the FIR filter.

In operation 950, the motion compensator 840 generates motion prediction blocks of the current block by performing motion compensation using the reference picture interpolated in sub-pixel units.

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, the spirit of the invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications will fall within the scope of the invention. In addition, the system according to the present invention can be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of the recording medium include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device and the like. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

Claims (17)

In the video encoding method,
Generating an extended corresponding block by extending a corresponding block of a reference picture used for motion compensation of a current block to a predetermined size;
Generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block;
Performing interpolation on a sub-pixel basis by applying a second filter to the filtered extended corresponding block; And
And performing motion prediction and compensation by using the reference picture interpolated in the sub-pixel units. The method of claim 1,
Generating the extended corresponding block
Determining a corresponding block of a reference picture most similar to the current block in integer pixel units; And
And extending the upper, lower, left, and right boundaries of the corresponding block by a predetermined pixel size to generate the extended corresponding block. The method of claim 1,
The first filter is an infinite impulse response (IIR) filter, and the second filter is a finite impulse response (FIR) filter. The method of claim 1,
Generating the filtered extended corresponding block
Changing a pixel value of integer pixels included in the expansion corresponding block by applying a first filter of a one-dimensional direction to the expansion corresponding block in a row direction; And
And changing a pixel value of integer pixels included in the extended corresponding block by applying a first filter in a column direction to the extended corresponding block having the modified integer pixel value. . The method of claim 1,
Generating the filtered extended corresponding block
The size of the extended corresponding block is NxM (N, M is an integer), z1 is a predetermined constant, and the pixel value of the kth integer (k is an integer from 0 to N-1) located in a predetermined row of the extended corresponding block. When s1 (k), the following equation; The value of c2 (k) calculated using c1 (k) = s1 (k) + z1 * c1 (k-1) and c2 (k) = z1 * (c2 (k + 1) -c1 (k)) Performing one-dimensional filtering in the row direction for the extended corresponding block by substituting the value of s1 (k) using; And
When the pixel value of the kth integer pixel (k is an integer from 0 to M-1) located in a predetermined column of the one-dimensionally filtered extended corresponding block in the row direction is s2 (k), the following equation; The value of c4 (k) calculated using c3 (k) = s2 (k) + z1 * c3 (k-1) and c4 (k) = z1 * (c4 (k + 1) -c3 (k)) And performing one-dimensional filtering in the column direction on the extended corresponding block by substituting the value of s2 (k) using. The method of claim 1,
Performing interpolation in units of subpixels
And performing interpolation on a sub-pixel basis using an FIR filter and an average value filter of a predetermined tap coefficient using a weighted sum of pixel values of integer pixels around the sub pixel to be interpolated. The method of claim 1,
And encoding predetermined index information indicating whether to predict and compensate for the motion using the filtered extended corresponding block. In the video decoding method,
Extracting motion vector information of the current block to be decoded from the received bitstream;
Generating an extended corresponding block by extending a corresponding block of a reference picture indicated by the motion vector to a predetermined size;
Generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block;
Performing interpolation on a sub-pixel basis by applying a second filter to the filtered extended corresponding block; And
And performing motion compensation using the reference picture interpolated in the sub-pixel units. The method of claim 8,
Generating the extended corresponding block
Determining a corresponding block of a reference picture most similar to the current block by using an integer pixel component of the motion vector; And
And extending the upper, lower, left, and right boundaries of the corresponding block by a predetermined pixel size to generate the extended corresponding block. The method of claim 8,
The first filter is an infinite impulse response (IIR) filter, and the second filter is a finite impulse response (FIR) filter. The method of claim 8,
Generating the filtered extended corresponding block
Changing a pixel value of integer pixels included in the expansion corresponding block by applying a first filter of a one-dimensional direction to the expansion corresponding block in a row direction; And
And changing a pixel value of integer pixels included in the extension corresponding block by applying a first filter in a column direction to the expansion corresponding block having the modified integer pixel value. . The method of claim 8,
Generating the filtered extended corresponding block
The size of the extended corresponding block is NxM (N, M is an integer), z1 is a predetermined constant, and the pixel value of the kth integer (k is an integer from 0 to N-1) located in a predetermined row of the extended corresponding block. When s1 (k), the following equation; The value of c2 (k) calculated using c1 (k) = s1 (k) + z1 * c1 (k-1) and c2 (k) = z1 * (c2 (k + 1) -c1 (k)) Performing one-dimensional filtering in the row direction for the extended corresponding block by substituting the value of s1 (k) using; And
When the pixel value of the kth integer pixel (k is an integer from 0 to M-1) located in a predetermined column of the one-dimensionally filtered extended corresponding block in the row direction is s2 (k), the following equation; The value of c4 (k) calculated using c3 (k) = s2 (k) + z1 * c3 (k-1) and c4 (k) = z1 * (c4 (k + 1) -c3 (k)) And performing one-dimensional filtering in the column direction on the extended corresponding block by substituting the value of s2 (k) using. The method of claim 8,
Performing interpolation in units of subpixels
And performing interpolation in units of subpixels using an FIR filter and an average value filter having a predetermined tap coefficient using a weighted sum of pixel values of integer pixels around the subpixel to be interpolated. In the video encoding apparatus,
A block extension unit for generating an extended corresponding block by extending a corresponding block of a reference picture used for motion compensation of the current block to a predetermined size;
A first filtering unit generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block;
A second filtering unit configured to perform interpolation on a subpixel basis by applying a predetermined second filter to the filtered extended corresponding block; And
And a motion prediction and compensation unit configured to perform motion prediction and compensation by using the reference picture interpolated in the sub-pixel units. In the video decoding apparatus,
An entropy decoding unit for extracting motion vector information of the current block to be decoded from the received bitstream;
A block extender configured to expand the corresponding block of the reference picture indicated by the motion vector to a predetermined size to generate an extended corresponding block;
A first filtering unit generating a filtered extended corresponding block by applying a predetermined first filter to the extended corresponding block;
A second filtering unit configured to perform interpolation on a subpixel basis by applying a predetermined second filter to the filtered extended corresponding block; And
And a motion compensator for performing motion compensation using the reference picture interpolated in the sub-pixel units. A computer-readable recording medium containing a program for implementing the video encoding method of any one of claims 1 to 7. A computer-readable recording medium containing a program for implementing the video decoding method according to any one of claims 8 to 13.

Images